vented explosion overpressures from combustion of...
TRANSCRIPT
![Page 1: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/1.jpg)
VENTED EXPLOSION OVERPRESSURES
FROM COMBUSTION OF HYDROGEN AND
HYDROCARBON MIXTURES
C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev
FM Global, Research Division, Norwood, MA, USA
3rd ICHS, September 16-18, 2009
![Page 2: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/2.jpg)
Background
• Venting is used to reduce the consequences of explosions
• The requirements for the venting have been specified in engineering standards
– NFPA 68 (2007)
– EN 14994:2007
• Standards based on empirical correlations of limited experiment data which may be off by an order of magnitude
• Hydrogen – Weak enclosures (rooms) – out of range of validity of correlations
– Strong enclosures (equipment) – based on questionable KG = 550
• There are other methods – V.Molkov – yet there are unresolved issues
Explosion Venting
![Page 3: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/3.jpg)
Background
• To generate a set of experimental data examining the effect of:– mixture composition– ignition location– vent size– obstacles – scale– …
• Use the experimental data to develop and validate a computational model
• Update technical recommendations and FM operating standards relevant to explosion hazards and to develop new models and engineering tools
Research Program Objectives
![Page 4: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/4.jpg)
Background
• Additional challenges
– Hydrodynamic flame instability is enhanced by thermal diffusion effects in lean hydrogen mixtures
– Effect of Le on the turbulent burning velocity
• Currently, these effects are not known well enough to be reliably modeled
• Comparisons with methane-air and propane-air mixtures should yield an insight on how to model them
Hydrogen Mixtures
![Page 5: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/5.jpg)
Background
• Examine the similarities and differences between three mixtures of similar laminar flame speed
– 18% hydrogen–air
– 9.5% methane-air and
– 4.0% propane-air
• Test an extension to the numerical CFD model developed in the previous studies and identify its capabilities and deficiencies to describe the physics responsible for the pressure build-up
Objectives of this study:
![Page 6: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/6.jpg)
Experimental Setup
• Overall size:
– 4.6 x 4.6 x 3.0 m
• Volume: – 64 m3
• Vent Sizes: – 5.4 m2 or 2.7 m2
• Vent Material:– 0.02 mm Polypropylene Sheet
Chamber Details
![Page 7: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/7.jpg)
4.6 m
4.6
m
Front-wall
Ignition
Back-wall
Ignition
Center
Ignition
Experimental SetupIgnition Locations:
Plan view:
3 Ignition locations
![Page 8: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/8.jpg)
Experimental Setup
• 3 mixtures, 3 ignition locations, 2 vent sizes and no obstacles
Test Parameters
6.47 ± 0.163.35.20.6418% Hydrogen
2.90 ± 0.142.97.50.389.5% Methane
3.31 ± 0.063.28.00.404.0% Propane
Average Measured Initial
Flame Speed, U0 (m/s)
Flame Speed,
(σ × SL) (m/s)
Expansion
Ratio, σLaminar Burning
Velocity, SL
(m/s)
Mixture
![Page 9: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/9.jpg)
0.0 0.2 0.4 0.6 0.8 1.0-0.05
0.00
0.05
0.10
0.15
Ov
erp
ress
ure
(b
ar)
Time (s)
Hydrogen 18.2%
Methane 9.5%
Propane 4.02%
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4-0.05
0.00
0.05
0.10
0.15
0.20
0.25
Ov
erp
ress
ure
(b
ar)
Time (s)
Hydrogen 18.1%
Methane 9.5%
Propane 4.01%
Experimental Results
• Explosion overpressures
Effect of Mixture composition
Center ignition 2.7 m2 vent Back-wall ignition, 5.4 m2 vent
![Page 10: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/10.jpg)
0 1 2 3 4 5 6 7 80
10
20
30
40
U/F
S
Position (m)
Hydrogen 18.2%
Methane 9.5%
Propane 4.02%
0 1 2 3 4 5 60
10
20
30
U/F
S
Position (m)
Hydrogen 17.9%
Methane 9.4%
Propane 4.01%
Experimental Results
• Flame speeds normalized by LFS
Effect of Mixture composition
Center ignition 2.7 m2 vent Back-wall ignition, 5.4 m2 vent
![Page 11: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/11.jpg)
0 1 2 3 4 5 6 70
5
10
15
20
U/U
0
Position (m)
Hydrogen 18.2%
Methane 9.5%
Propane 4.02%
0 1 2 3 4 5 60
5
10
15
U/U
0
Position (m)
Hydrogen 17.8%
Methane 9.4%
Propane 4.01%
Experimental Results
• Flame speeds normalized by initial flame speeds
• Enhancement of the hydrogen propagation speed caused by flame instabilities remains constant throughout the combustion process
Effect of Mixture composition
Center ignition 2.7 m2 vent Back-wall ignition, 5.4 m2 vent
![Page 12: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/12.jpg)
Numerical Model
• OpenFOAM (Weller et al. 1998)
– Open source Field Operations And Manipulation
• Solver details
– Fully compressible implicit NS solver
– 2nd order discretization schemes in time and space
• LES model
– One equation eddy viscosity model for sub-grid turbulence
OpenFOAM
![Page 13: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/13.jpg)
Numerical Model
• Regress Variable Combustion Model
• Sub-grid flame wrinkling, Ξ, is due to both turbulence and flame instabilities
• Taylor instability model important to resolve external explosion (ICDERS-2009)
• Separate transport equations for ΞRT
and ΞT
– Assumption of different dominant length scales
Partially Pre-Mixed Combustion Model
( ) ( ) bΞSbbt
bLu
~~~~~
∇−=∇⋅∇−⋅∇+∂
∂ρρρ
ρDU
TRTHI ΞΞΞΞ ⋅⋅=
![Page 14: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/14.jpg)
Numerical Model
• Flame surface area increase due to hydrodynamic instability
• Constant for given grid size (∆) and given mixture (λc)
• θ(hydrogen)/θ(methane, propane) ≈ 2.4 – consistent with known values of λ
c
Combustion Model - HI
3/13/13/1
0
∆
∆=
=
c
m
c
m
A
A
λ
λ
λ
λ
[ ]θλ
,1max,1max
3/1
=
∆=
c
IHI aΞ
![Page 15: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/15.jpg)
Numerical Model
• Sub-grid wrinkling due to Turbulence (Weller et al 1998, Bradley et al 1992)
• All mixtures: αT
= 0.7
• Factor Le-n undistinguishable from ΞI (n = 0.5)
Combustion Model - Turbulence
( ) ( ) TstTTTsT
ΞΞRΞGΞt
Ξσσ −−−−=∇⋅+
∂
∂1U
ητ
28.0=G
1−=
eq
eq
Ξ
ΞGR
( ) ( ) n
LTeq LeSuaΞ−∆= 6
1
2
1' // δ
![Page 16: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/16.jpg)
Numerical Model
• Unstructured
Grid
• 0.05m cell size
• 1.2M cells
Computational Grid
ChamberVent
![Page 17: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/17.jpg)
Results – Simulations
Back Ign. 5.4 m2
vent 0 obs.:
0.0 0.1 0.2 0.3 0.4 0.5 0.6
-0.04
0.00
0.04
0.08
0.12
0.16
Ov
erpre
ssu
re (
bar
)
Time (s)
18.5% Hydrogen
18.2% Hydrogen
Simulation
Center Ign. 5.4 m2
vent 0 obs.:
0.0 0.1 0.2 0.3 0.4 0.5
0.00
0.02
0.04
0.06
0.08
Ov
erp
ress
ure
(b
ar)
Time (s)
18.2% Hydrogen
Simulation
![Page 18: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/18.jpg)
Results – PropaneBack Ign. 5.4 m2 vent:
0.0 0.2 0.4 0.6 0.8 1.0 1.2
0.00
0.02
0.04
0.06
Ov
erp
ress
ure
(b
ar)
Time (s)
4.0% Propane
4.0% Propane
Simulation
![Page 19: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/19.jpg)
Results – SimulationsMovie
![Page 20: VENTED EXPLOSION OVERPRESSURES FROM COMBUSTION OF …conference.ing.unipi.it/.../stories/presentations/ID176-Dorofeev.pdf · C. Regis Bauwens, Jeff Chaffee and Sergey Dorofeev FM](https://reader033.vdocuments.site/reader033/viewer/2022043007/5f93d76fb5f1fc62a2791b11/html5/thumbnails/20.jpg)
Conclusions• Experiments showed that flame speeds and overpressures
in H2 mixtures were much higher than that in methane and propane due to flame instabilities, despite close laminar values
• Laminar flame speeds are not sufficient to characterize mixture reactivity in vent-sizing for hydrogen mixtures
• CFD model was tested that takes into account mixture properties, flame instabilities and turbulence
• Numerical results reproduce basic features observed in experiments, such as overpressures and flame speeds for the range of parameters studied
• Further studies are planned to include effects of scale and obstacles in the model validation exercises